Rolling out Epic across an organization the size of St. Luke’s isn’t a project; it’s a journey that requires an enormous effort to stay on course. It also requires a lot of change, and when that becomes too much, leaders must be willing to hit pause. It’s precisely what CIO Marc Chasin did three years ago, and the result was a more standardized, collaborative approach. In this interview, he talks about how his team is looking to optimize and stabilize its EHR, the federated approach they’ve adopted with data warehousing, and the ultimate goal with patient engagement. Chasin also discusses the application rationalization process, the two types of CIOs we’ll see going forward, and what he hopes to accomplish as a member of CHIME’s Board.
- About St. Luke’s
- Fully on Epic since 10/2016 — “We’re in stabilization mode.”
- 6-year journey with a year-long hiatus
- “The organization had hit their threshold for change.”
- Dealing with detractors — “We had to manage the message.”
- A federated EDW model
- Managing at-risk lives—“We have to learn how to be competitors and partners.”
- Focus on data governance
LISTEN NOW USING THE PLAYER BELOW OR CLICK HERE TO SUBSCRIBE TO OUR iTUNES PODCAST FEED
This installation, while a heavy lift, has been hugely successful for our organization — and not only for the deployment of technology. We really are starting to function as a health system and realize that we are more similar than different.
We essentially put those providers on an island. They were communicating very effectively, as long as the patient was in the ambulatory setting and seeing specialists and primary care doctors. As soon as they had to cross the threshold into the hospital, we still had the same problems.
Communication is key and imperative, and it’s a competency that CIOs don’t focus on. I’ve learned this is one of the things I have to focus on, because if I can communicate things succinctly, people will buy in; people will come along on the journey.
We’re looking at a federated model in which we don’t have to consolidate and have one large data warehouse. We should be able to pick and choose based upon the question our end user is asking, and serve that information up in the most appropriate way possible.
If you get poor data capture, the information you’re trying to glean on the other end is going to be off by a magnitude. We want to make sure that our data stewards, as well as our end users, have the highest confidence in the information that they are receiving.
Gamble: Hi Marc, thanks for taking some time to speak with us today.
Chasin: Thank you.
Gamble: I know we spoke about three years ago, but can you give a high-level view of St. Luke’s Health System — where you’re located, number of hospitals, things like that?
Chasin: Sure. St. Luke’s Health System is located in Boise, Idaho — at least the corporate office is. We are the largest health system in the state. We have eight acute care hospitals, and over 450 employed physicians that span the lower southwestern part of the state.
Gamble: The last time we spoke, you were in the thick of implementing Epic. What’s the status is now? Is it implemented in all of the hospitals?
Chasin: Sure. As of October 1, 2016, we are completely installed through our inpatient and our ambulatory environment — the entire Epic suite. Additionally, 60 days post-install we achieved HIMSS Stage 6 designation on both our inpatient and our ambulatory environment.
Gamble: As we had talked about, the hospitals had previously were on different systems; so it was really a big initiative. Now, especially since achieving HIMSS Stage 6, are you on optimization mode at this point?
Chasin: We are still in stabilization. We have identified some key workflows that have to be relooked at, and some areas of our facility that need some really focused detail. Namely, we’re looking at the NICU and really optimizing and stabilizing in that area, and as well as our interventional radiology areas. For the most part though, Epic has been a huge success. Our patients are being treated very well. Information is flowing throughout the continuum of care, and we feel that this partnership and this installation, while a heavy lift, has been hugely successful for our organization — and not only for the deployment of technology. We really are starting to function as a health system and realize that we are more similar than different, and that standardization ultimately will drive toward the improvement of care and reduction of cost.
Gamble: Right, the ultimate goals. Now obviously, there’s no secret sauce; but what do you think have been some of the keys to really establishing that integration across the system and functioning as one health system when you have these very different facilities?
Chasin: That’s a very good question. I don’t know if I have a specific answer. I could tell you from our experience that this was a six-year journey, and I think we matured as an organization throughout those six years. Three years ago when we had spoken, we ended up taking a year hiatus from install. I felt that the organization had hit their threshold for the amount of change that we were putting upon it, so we took a year off and relooked at our governance. We looked at why we’re doing what we do and the mission and vision of our organization, which to serve our patients. We relooked at that and we came around the table and said, ‘This is what we need to do. We need to complete this project in order to really treat our patients the most appropriately and the best we can.’
In redoing that governance, we brought district hospitals and clinics together and worked really hard to show that we all have the same vision and the same goal of treating the patients of our community as best as we can. I think that pause and the maturation of our organization drove us toward a more standardized collaborative approach throughout our system.
Gamble: Really interesting. When you did take that pause, were there hospitals that were at different stages? I imagine the timing was something that had to have been worked out.
Chasin: Yeah, the timing had to be worked out. Fortunately, we had just finished our ambulatory installation, so the majority of our physicians were on Epic just in the ambulatory. What we had to contend with, though, is that we essentially put those providers on an island. They were communicating very effectively, as long as the patient was in the ambulatory setting and seeing specialists and primary care doctors. As soon as they had to cross the threshold into the hospital, we still had the same problem of a break in the care and the inpatient facilities searching for information.
But we only had to deal with that for two years. As of October 1, 2016, we’re live in the entire system. So while it was difficult for those two years, I think it was well worth it. While we got our governance set up, we started to come together as a health system and understand our shared mission for the patients of our region.
Gamble: When you do have a strategy like that, I imagine one difficulty might be that on the leadership side, you want to just get this done. You just want to get to that point of implementation. So I’m sure that it wasn’t the easiest thing to do, but I guess the pros outweighed the cons.
Chasin: Yeah. We had to message this appropriately. In any project of this magnitude, you do have your detractors. Many of the detractors did not feel that we were ever going to go live or we would never make this conversion. The individuals — both providers and administrators — who were really engaged felt like, let’s do it right as opposed to let’s do it quickly. We had to manage the message very well.
What I’ve found these past few years is that communication is key and imperative, and it’s a competency that CIOs don’t focus on. I’ve learned that this is one of the things that I have to focus on, because if I can communicate things succinctly, people will buy in; people will come along on the journey. It’s where there’s gaps in communication and gaps in understanding when the negative starts to fill that void.
Gamble: In terms of the data warehouse, is that something where you also went with Epic? What has been the strategy there?
Chasin: That is a very interesting question that we could probably speak for a couple of hours on. We have gone with the Caboodle/Cogito data warehouse. We do have our own enterprise data warehouse, and we have multiple other applications that have their many data warehouses. We are in the process of looking at our strategy from an advanced analytics and machine-learning perspective. We’re looking at a federated model in which we don’t have to consolidate and have one large data warehouse. We should be able to pick and choose based upon the question our end user is asking, and serve that information up in the most appropriate way possible.
Strictly speaking, we aren’t actively populating the Caboodle/Cogito data warehouse, but we’re looking at it right now to answer event-level questions that reside within the Epic record, whether they have to do with patient registry, patient care — small populations of patients. If we’re looking at how we’re going to drive care in our region from a cost and quality perspective, we need to bring in external data sources, and we need to bring in our cost data. We use our federated enterprise data warehouse in that skewed-out lens in order to answer questions like that.
Gamble: Yeah, that sounds like that would be a good discussion for another time because there’s a lot there.
Chasin: There sure is.
Gamble: Now, when you talk about the bigger goals of improving health not just locally but really throughout the entire system through things like accountable care and population health, what is the strategy at this point?
Chasin: Sure. We’ve spent a fair amount of time on this within our health system and our leadership. We have developed St. Luke’s Health Partners Network, which takes a bigger look at how we treat populations. Today, we are responsible for 170,000 lives within the southwestern part of Idaho. That equates to almost a billion dollars in revenue, and we’re at total risk for that as of January 1. We need to, as a health system, learn how to become much more efficient in the way we deliver care and much more astute as to the needs of the specific populations, because not all those patients are seen by St. Luke’s providers. Our network spans some of our competitors and some of our partners, and we have to learn how to be competitors and how to be partners in the same region. We are using our data to learn where we experience risk, and how we segment our populations to get them more personalized care to achieve the lowest total cost of care within the highest quality parameters that we can deliver.
Gamble: You just touched on it, but that data governance component is so huge. Can you talk about your strategy for having that effective governance model?
Chasin: I would say that this is where we’re working really intently on data governance. It is very difficult for many of my leaders and my partners to really comprehend how to manage data and to govern data when it’s something that’s not really tangible. Many of our data definitions are provided to us by CMS, but there are others in the financial area that we have to come around and understand who owns the data, who’s the data steward and who is managing that area, and hold them accountable to make sure that that process is maintained and it is solid.
Because as you know, if you get poor data capture, the information you’re trying to glean on the other end is going to be off by a magnitude. We want to make sure that our data stewards as well as our end users have the highest confidence in the information that they are receiving so they can make rapid improvements to the delivery model that we’re deploying here at St. Luke’s.
Gamble: That’s a really interesting concept. You said the patient population that you serve also sees other providers that are outside of St. Luke’s system. What has been the level of exchange with those organizations?
Chasin: I’m glad you brought that up because interoperability is a very keen interest of mine. I’ve spent a fair amount of time over the past couple of years with the Sequoia Project doing the Trust Framework as well as with Epic on the Care Everywhere Network. We have tried to develop any and every opportunity for a partner to connect with us. We send CCDs. We are active on the Sequoia Project, so we are connected to the eHealth Exchange. We have our own local HISP that we offer to connect to our providers as well as the local or state-run Health Data Exchange.
So we have really good partnerships in the exchange of data and CCDs, and we’re always looking to improve that to get it much more robust, and to get the query and the record locator service as streamlined as possible. We believe that the exchange of patient information should be transparent, and where the patient lands, their information should be that way. That’s the vision that we are going with here at St. Luke’s.
Gamble: St. Luke’s is part of Epic Everywhere?
Chasin: Yes, we’re part of Care Everywhere. We’re also part of Sequoia Project, which is now Carequality.
Gamble: Being in a rural area, can you talk about what kind of progress the organization is seeing with telemedicine and what barriers are still out there?
Chasin: Sure. You’re correct in that St. Luke’s Health System serve around 43,000 square miles of Southwestern Idaho and many of that is rural in nature. We do have some critical access hospitals in those areas, but there are still gaps in specialty care that the city or the tertiary centers need to provide.
We have a very robust tele-ICU program in which we offer intensivist service to these remote areas. We believe that the best care is care as close to home as possible. And if we can provide those services, we feel that patients are going to get the right care at the right time, they’re going to feel more comfortable near their family members, and they’ll get better quicker.
We offer one of the largest teleICU services in the northwest, but that’s only the beginning. We’re starting to develop teleneurology capabilities, and now going forward, as we have to start managing this population that we’re completely and totally at risk for, we want to deploy telehealth services into the homes, into the workplace, so patients can receive care where you want to as opposed to always having to come to the doctor. We’re not there yet, but that is on our roadmap.